The project proposed in this application is a response to PAR-02-062, "Building Translational Research in Behavioral Science." The overarching purpose of the proposed project is that described in PAR-02-062, to "increase collaborative efforts among basic and clinical researchers in order to realize the fullest potential of translational research," for the particular purpose of studying "decision making as it affects the behavior of patients and providers in mental health or drug abuse services settings." This project brings together researchers in psychiatric rehabilitation, experts in mathematical modeling with backgrounds in engineering, computer science and psychology, a psychologist researcher in physical rehabilitation, an experimental psychologist who specializes in human perception and processing of complex data, and an expert in consumer interests and perspectives. These collaborators will work together to develop mathematical models of progress through rehabilitation, working from existing and planned clinical databases. The models will be developed with special attention to identifying key decision points in the rehabilitation process, and identifying patterns of circumstances and consumer characteristics that should inform clinical decision making by providers and consumers. When these patterns are identified, the findings will be translated into informatic design parameters, to inform future development of clinical decision support systems for rehabilitation. Advanced decision support systems, based on the findings of the modeling analyses, are expected to greatly enhance the outcome of psychiatric rehabilitation by alerting practitioners and consumers to make key judgements and decisions, and by providing the information they need to make those decisions. The project proposed in this application is a key step in developing the interdisciplinary collaboration necessary to produce advanced, research-informed clinical informatic systems for psychiatric rehabilitation.